INDEX

Explanations

phrases related to strong or impactful actions or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "strike" and its variations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iliary

-0.73

umption

-0.72

ilon

-0.69

ŀ

-0.68

itive

-0.68

ended

-0.67

adian

-0.66

HUD

-0.66

ORY

-0.66

ITY

-0.64

POSITIVE LOGITS

 resemb

0.88

ched

0.75

 resemblance

0.74

tle

0.74

breakers

0.72

 striking

0.71

ting

0.71

tered

0.71

 similarities

0.71

achi

0.68

Activations Density 0.041%

phrases related to strong or impactful actions or events

occurrences of the word "strike" and its variations

No Comments

No Known Activations